We sought to establish reference values for a new direct assay for small dense LDL cholesterol (sdLDL-C) and to measure sdLDL-C concentrations in patients with established coronary heart disease (CHD) vs controls. Direct LDL-C and sdLDL-C were measured in samples from 3188 male and female participants of the Framingham Offspring Study, including 173 men and 74 women with CHD. Postmenopausal status and male sex were associated with higher sdLDL-C concentrations (P < 0.0001). Cholesterol-lowering medication use was more frequent (P < 0.0001) in CHD patients than in controls (46.8% vs 11.4% in men; 35.1% vs 8.8% in women). In men, mean LDL-C was lower in CHD than in controls (3.22 vs 3.51 mmol/L, P < 0.0001), whereas mean sdLDL-C concentrations were similar (0.83 vs 0.84 mmol/L, P = 0.609). In women, mean LDL-C was similar in CHD and controls (3.53 vs 3.46 mmol/L, P = 0.543), but mean sdLDL-C was higher (0.83 vs 0.68 mmol/L, P = 0.0015). The mean percentage of LDL-C as sdLDL-C was higher in both men and women with CHD than controls (P < 0.01). Increased LDL-C and sdLDL-C were found in 10.4% and 22.0% of men and in 24.3% and 27.8% of women with CHD, respectively. Despite 4-fold greater cholesterol-lowering therapy use, CHD patients had mean LDL-C concentrations above the LDL-C goal of <2.6 mmol/L (<100 mg/dL). Although women with CHD had higher sdLDL-C concentrations than controls, this difference was not seen in men. These findings may explain some of the high residual risk of future CHD events in CHD patients.